zlacker

How is he exaggerating the situation? What is false about the criticism? Are you referring to a previous time where they cried wolf? I read through the Twitter thread and GrapheneOS seemed pretty even keeled and above board about it to me (even if that is uncharacteristic)

>>oharap+(OP)
Graphene's claim of "AOSP is dead" is easily verifiable.

> This also marks the availability of the source code at the Android Open Source Project (AOSP). You can examine the source code for a deeper understanding of how Android works, and our focus on compatibility means that you can leverage your app development skills in Android Studio with Jetpack Compose to create applications that thrive across the entire ecosystem.

https://android-developers.googleblog.com/2025/06/android-16...

This was posted 2 days back.

>>oharap+(OP)
He is not exaggerating the situation he is lying. There is no basis for his very clear and serious claim that AOSP is dead.

>>bitpus+Z1
But you don't have any ability to run AOSP on any devices? Free as in literally unrunnable?

AOSP feels incomplete without there being some flagship way to use it.

>>bitpus+Z1
If you want to define "AOSP is not dead" as "there exists a source-available AOSP repo that is not ground-up buildable for any real world device without losing major features like SecureBoot", that's fine, but that's not the definition being discussed.

Absent device trees, AOSP as of the Android 16 release is a subset of the utility of Android 15. If one sees the use of AOSP as mainly relying on the now absent functionality, then declaring "AOSP is dead" is not unreasonable.

If the Linux Foundation sold itself to Microsoft, ceased publishing kernel sources or binaries, and declared henceforth Linux would exist as WSL and nowhere else, it would be reasonable to say "Linux is dead" even if something with a subset of that functionality, named "Linux", still existed.

>>jaunty+Ni
It is the literal source code. You can read it, study it, distribute it, modify it. And I'm pretty sure the license allows you to sell it.

That's really really permissive.

>>bitpus+VB
They are talking about the device tree which has not been released with Android 16. So no way to run a custom kernel on actual hardware, unless that hardware is open. AOSP is a car without an engine now.

>>bitpus+Z1
That's "source available" not actual open source. If you can't build it and run it, you can't verify that it builds correctly.

>>bitpus+VB
I guess I can also paper my walls with it...

>>margal+Dv
> Absent device trees, AOSP as of the Android 16 release is a subset of the utility of Android 15. If one sees the use of AOSP as mainly relying on the now absent functionality, then declaring "AOSP is dead" is not unreasonable.

There are a million devices out there that build on AOSP that are not Google Pixel. This is a Pixel news, not AOSP news.

>>tgma+UZ
And every single one of those devices is missing features of AOSP, notably secure boot.

Google pixels were until recently the only phones able to run AOSP with 1:1 feature parity. And now there are none.

>>paxys+N4
How would you more accurately characterize current situation with Android 16?

>>jaunty+Ni
Having stewed on this for a week, I'm even more morose over this.

I'm getting particularly salty that this is happening exactly as Android hurdles two huge integration challenges, as it goes from a standalone not-Linux-desktop single-screen computing device to something vastly more: a multi-screen capable, virtualized Linux desktop running device. Two huge leaps of integration.

This is just a maddening maddeningly crucial leap forward that Android is making right now, and it's woeful beyond words to see it making such a bold leap but leaving open-source totally behind at this exact junction, where the OS actually integrates with the hardware reasonably well/with more than the most trivial complexity for the first time ever.

This is just such a shitty shitty shitty turn of events.